Short-Term Reversal Strategy Still Outperforms The Market
Summary
- The short-term reversal anomaly in stocks has persisted over the past 20 years.
- A simple rotational strategy can be used by investors to benefit from this stock market anomaly.
- A historical backtest of this strategy offers insight in the risk/reward profile of the strategy.
As the stock market is currently rallying up again after the COVID-19 correction, investors are analyzing whether now is a good time to make an investment in stocks.
Unlike passive investors, who prefer to invest in large stock market trackers, certain investors prefer a more active approach.
In this article, we will have a look at a stock market anomaly called the "short-term reversal strategy", in which funds are being allocated into stocks which have declined strongly over a short period of time. The positions in these stocks are then closed after a brief period of time and the funds are allocated again to the weakest performers.
By basing investment rules systematically on this stock market anomaly, investors would have outperformed the stock market significantly. We will go through the results of this strategy and present a simple rotational strategy which can be implemented by traders to benefit directly from this anomaly.
Historical research of the reversal anomaly
Short-term return reversal in the stock market is a well-established phenomenon, which has been researched in depth by academicals. For example:
- Jegadeesh (1990) documented profits of about 2% per month (or 24% annually) over the period 1934-1987 by using a reversal strategy buys the worst-performing stocks over the past month and holds these stocks in the portfolio for one full month.
- DeBondt and Thaler (1985) look in-depth at this phenomenon and argue the reversal effect can be explained by over-reactions to information in the market
- De Groot et al (2011) investigate a simple mean reversion strategy in US stocks (based on the anomaly) and propose the implementation of a more sophisticated portfolio construction algorithm to lower the turnover and trading costs of the strategy.
While most of the historical research tends to agree on the existence of the short-term reversal anomaly, they point out the practical issues of implementing this strategy due to the high turnover which will generate material trading fees. These trading fees need to be reduced from the gross profit of the trading strategy in order to define the net profitability.
The last study (2011) mentioned here above is the most relevant for us, as this analysis looks in detail at the trading costs of the strategy and proposes two measures to limit the total trading costs:
1) Focus on investable, large-capitalization stocks only;
2) Deploy a slightly more advanced rebalancing rule;
A simple rotational strategy
Based on the academic findings on the short-term reversal anomaly, we will now construct a simple mean-reverting trading strategy ourselves. The goal is to investigate whether we can buy and sell the worst-performing stocks and generate a net profit after deducting the transaction costs.
In order to do this, we will backtest a trading strategy which the following characteristics:
- Stock data: We will only invest in stocks that are part of the S&P 100 index. The stocks within this index are all large-cap stocks and high in liquidity. This rule is in line with the recommendation of the research of De Groot et al (2011).
- Testing period: 1 January 2000 - 30 April 2020. This large time span will ensure we can analyze the results of the strategy during bull and bear markets.
- Starting equity: $10,000.
- Investment in each stock: 20% of the available equity (we equal size all our holdings when the purchase is made).
- Maximum # stocks in portfolio: 5
- Buying rule:
- Each Friday evening after the markets have closed we will calculate the 5-day performance of each stock in the S&P 100 index.
- Based on this performance, we will rank the stocks from the worst performance to the best performance.
- The worst 5 performers will be purchased on the next Monday at their respective closing price.
- Selling rule: Once a stock has been purchased it will be held in the portfolio until it no longer belongs to the top 50 worst performers of the S&P 100 index. If the stock no longer belongs to the worst 50 performers, it will be sold on the next Monday at the respective closing price.
- Commissions: $1 fee will be charged each time a stock is purchased or sold.
- The platform used: Amibroker
Example of stock selection
For example, on 7 January 2000 Goldman Sachs (GS) closed at $82.5 which was 12.3% below of its closing price on 31 December 1999. This stock was one of the 5 worst-performing stocks on 7 January 2000. We would purchase the stocks at the closing price of $84.3 on 10 January 2000.
On Friday 14 January 2000 we see GS is no longer one of the 50 worst performers so the position will be closed by selling the GS stock at the closing price of the next Monday for $86.8 (a net profit of 2.86%). To replace the GS stock, we would then purchase AMZN stock (one of the worst performers in that specific week).
In the table here below we give an overview of the trades which would have been made between January 2000 and mid-March 2000.
Source: Backtest results created by the author
Results of the backtest
In the section below we discuss the results of the backtest.
The total starting equity of $10,000 would have grown up to $187,580 (x18 folded). We would have executed 2,395 trades of which 61% would have been winners. The annualized return would have been a whopping 15.5%.
Source: Backtest results created by the author
In the chart below we can see the equity curve which would have been the result of following our trading system:
As we can see from the chart the equity has appreciated significantly throughout the past 20 years, ending at $187k. If we would have invested $10k in a large stock market tracker (SPY) the ending value would only be around the $30k level.
If you look at the equity here above closely, you can see the current equity value has been much higher only a few weeks ago ($300k at its highest high). Indeed, the short-term reversal system tends to outperform in the long run, but it has volatile swings throughout its lifetime. In the chart below we present the underwater equity curve, i.e. the dives your equity would have taken from its highest high levels:
In the table here below we present the monthly returns:
To recap the results of the backtest:
- The strategy results in a large number of weekly trades, of which 60% are winners.
- The equity curve of the strategy is mostly up-trending and outperforms a passive investment in the general stock market (SPY)
- Even after deducting commission fees, the results are profitable.
- The strategy is volatile and vulnerable to sharp stock market corrections. In March 2020 for example, the strategy lost 39.3%.
- In the long run, these negative months have been compensated by following positive months, making the strategy profitable as a whole.
Practical Application & Risks
In order to bring all of the information from the historical research and our backtests together, we can now address the question: Can traders still use this strategy as of May 2020 and benefit for the years to come?
As academic research has pointed out, there is obviously an outperformance effect present here. By using the short-term reversal anomaly we can detect the worst-performing stocks which on average tend to outperform the market in the coming weeks. The backtest we presented in this article would have 18-folded the initial investment, whereas a passive investment in the SPY tracker would have "only" tripled your initial investment.
In this regard, it makes sense for active traders to use this rotation strategy to actively trade these stocks and aim to outperform the stock market. The stock market is currently rebounding from the low levels it reached following the COVID-19 correction, backed by the quantitative easing program recently launched by the Federal Reserve. Assuming the S&P 500 index will reach it previous high levels and continue the long-term uptrend in stocks, we do expect the short-term reversal strategy to outperform a simple buy and hold strategy in the SPY.
The strategy has its own risk profile though, as we can see from the results from our backtest. As the rotation strategy is designed around buying the worst-performing stocks of the S&P 100 Index, we will from time to time buy stocks which will decline even further from this point on. In the long-run though, this underperformance is balanced out again by the results of stocks who reverse to the upside after their short-term price decline. The same is true for the transaction costs, as we will have a high turnover each year, it is important to execute the strategy with a stockbroker with competitive rates. In this backtest, we used a fixed fee of $1 per transaction to include these costs.
Conclusion
In this article, we explained a well-known stock market anomaly, the short-term reversal effect. This effect demonstrates how stocks with short-term negative price performance tend to outperform the stock market in the following period.
In the second section, we present a simple mean-reversion strategy with trading rules-based to purchase the worst-performing stocks. This strategy will be tested on 100 large-cap stocks and the results are discussed in the last section.
In the third section, we briefly discuss the practical application of this strategy and the risks involved.
The results demonstrate the short-term reversal effect is still present in stocks, as the worst-performing stocks tend to outperform the stock market. Even after deducting commission fees for each trades, the results are significant. The trading strategy does come with material downside risk, as some of the worst-performing stocks tend to decline even further. In the long run, this effect is compensated again by the stock performance of our winners, but in the mid-term, we can expect large equity drawdowns.
This article was written by
Analyst’s Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.
Recommended For You
Comments (14)
Please let me know in case you would have any additional question.
